Implications for economic environmental valuation

After having developed a novel framework on the basis of an individualistic economic approach that is nevertheless able to explicitly address social values, we now turn to the implications of including social values into the consideration for economic environmental valuation.

The above discussed value pluralism and multidimensionality of motivations behind preferences is not adequately reflected by the current conception of TEV. For non-use values the classification is rather imprecise and does not allow for value justifications beyond instrumentalism nor can it capture values associated with human embeddedness in nature.

Therefore, Figure 4-6 illustrates an extended and refined TEV in order to account for the complexity in terms of value dimensions and motivations (for the original TEV concept see Figure 2-3). The differences between the “traditional” TEV and the extended version associated with social values (Figure 4-6) will be described in the following.

Firstly, the extended framework explicitly accounts for the fact that ES and economic environmental valuation are inherently anthropocentric concepts and hence, following the valuation’s subjectivist approach only anthropocentric values can be accounted for. Secondly, the extended framework distinguishes between instrumental and non-instrumental values. As discussed above, this is relevant because altruism may either be linked to a human valuer’s utility (sympathy) or be independent of the latter (commitment). Altruism motivated by sympathy is here designated as warm glow value, whereas altruistic value associated with commitment is referred to as common good value. The latter refers to what is beneficial to other members of a community/society but independent of the valuer’s personal welfare. As outlined in Figure 4-6, warm glow value may relate to nature and non-human entities (existence value) or humans (philanthropic value). Bequest value is depicted as special form of altruism or rather warm glow towards future generations. Existence value, bequest value and intragenerational altruism based on sympathy are all captured by anthropocentric instrumental value as it is assumed that the person valuing obtains satisfaction herself from these types of values. While various definitions of existence value exist, it is often understood as special form of altruism (see e.g. Turner, 1999), motivated by sympathy towards non-human entities, and hence in the

original framework associated with ‘altruism to nature’ (see e.g. TEEB, 2010, p. 195). This definition is also adopted in the TEV concept presented here.

However, this concept of existence value does not account for the complex human-nature relationships and connectedness as discussed above, criticism generally voiced in the context of the ES framework (see Chan et al., 2012a). In response to this criticism and to account for value pluralism, weak anthropocentric intrinsic value is included as third overarching value category besides anthropocentric instrumental value and anthropocentric non-instrumental value. At a first glance, the label of weak anthropocentric intrinsic appears to be an oxymoron, it shall be illustrated why it is an adequate description. In a strict sense anthropocentrism is the belief that value is only associated with humans implying that non-human entities are only means to human ends. In its “weak” form anthropocentrism does not imply that nature does not

Total Economic Value

Source: Own illustration based on TEEB (2010, p. 195)

Colours indicate if values are associated with anthropocentric instrumental value (white), anthropocentric non-instrumental value (light grey), or weak anthropocentric intrinsic value (dark grey).

Figure 4-6 Extended total economic value framework to account for value pluralism and multidimensional motivations

have value in itself (intrinsic value) but simply that the act of humans valuing is always anthropocentric, even if they try to adopt the view of a non-human entity (Callicott, 1989;

Hargrove, 1992), but not necessarily instrumental. Hence, weak anthropocentric intrinsic value is not an oxymoron but suggests that humans value a non-human entity for itself. That is to say, from an anthropocentric point of view the non-human entity is valued as an end instead of a mean to an end (Hargrove, 1992). Incorporating this category of value does not overcome the criticism that objective intrinsic value of nature cannot be incorporated in economic environmental valuation.¹³ Yet, it broadens the scope of values and allows for more complex preferences and human behaviour. Importantly, this category of intrinsic value is not only associated with a deontological sense, meaning ascribing existence rights towards non-humans and duties towards species or ecosystems (deontological biocentrism), but also understood in the context of sense of connectedness and related concepts of caring and stewardship (virtue biocentrism) (see Palmer, 2017). To make this argument of virtue, caring and human-nature relationships distinct, this category is further specified as biocentric and biophilic value.

Further, as indicated by the arrow at the bottom of Figure 4-6, these values as well as anthropocentric non-instrumental values are assumed to be self-transcendent. In contrast moving towards the left, values become more and more focussed on individual interests and use. Hence, these values are linked to self-enhancement with reference to Schwartz’s theory of basic values (see Schwartz et al., 2012).

Conceptualising intrinsic value in a deontological sense implies moral rights which are in conflict with welfarism and generally, consequentialism. The same applies to the typology of non-instrumental common good value. Hence, these value typologies are also conflicting with conventional economic valuation due to violation of the axioms described in Chapter 2. In contrast, consequentialist biocentrism values states of affairs instead of non-human entities as ends in themselves. Simply put, well-being of (living) non-human entities is valued instead of nature in itself (Attfield, 1995; Attfield, 1998). However, this does not suggest that all living things are treated equally (Attfield, 2005). All three forms of biocentrism are assumed to be action-guiding (Palmer, 2017). Hence, they are relevant for economic valuation. Generally, economic valuation methods lack the ability to uncover motivations behind value expressions.

13 Cf. Ward (1999) who argues that objective value can be incorporated into economic environmental valuation based on Harsanyi’s utilitarianism. Following Goodin’s idea that interests originate from objective value (Goodin, 1996), Ward (1999) argues that utility functions can be ascribed to non-human entities through humans empathising with them. In this case, utility is not understood in terms of pleasure and pain but refers to interest and risk as in Harsanyi’s expected utility maximisation approach.

Accordingly, conventional valuation methods and the associated market context have been criticised for being unable to meet this complexity. Ignoring plural values, multiple preference orderings and/or preference endogeneity may result in double-counting or undervaluation (Blamey et al., 1995; Brekke & Howarth, 2000). Further, adequacy of self-interested individual preferences in a social context and in case of value pluralism have been questioned.

It has been illustrated that the valuation context may influence value expressions. The context shapes the form of interaction by influencing the perception of the object as well as which transcendental values, beliefs, etc. are considered to be relevant which ultimately results in the expression of assigned value (Brown & Slovic, 1988). Hence, economic environmental valuation is considered to represent a VAI which affects the type of knowledge and knowledge construction, as well as perception and type of motivation or rationality (Jacobs, 1997; Vatn, 2017). The institutional context defines the rules of the game, facilitating certain rationalities while reflexive humans still have to interpret the context (Hansjürgens et al., 2017; Vatn, 2017).

These contextual effects are especially relevant for complex public goods as people usually lack experience in paying for them because these goods are usually not paid for in real life. Hence, hypothetical markets have to be designed. Due to lack of experience people often do not have well-established preferences which makes them more sensitive towards contextual effects (Brown & Slovic, 1988). This poses the question what value expressions may potentially be elicited through economic valuation methods and additionally, when deviations from individual rationality in form of social value expressions become relevant.

As discussed above, social values are a broad concept associated with value pluralism, hierarchical values, multidimensionality of motivations, social influences and interdependencies. In valuation contexts social values can be identified based on the value provider, value scale, intention, and/or elicitation process (Brown, 1984; Hansjürgens et al., 2017; Kenter et al., 2015). The value provider may either be an individual as in conventional methods, an individual in a group context, e.g. a focus group followed by elicitation of individual WTP, or a group agreeing on a WTP. In line, the scale may refer to the individual, groups or society. As discussed above, intention can be self-regarding (I-preferences), other-regarding (We-Preferences) or impartial and impersonal (We-Preferences). In the following, various elicitation processes are considered which either elicit stated preferences individually, in a social context (potentially) characterised by a social learning process or behind a veil of ignorance. On the basis of four relevant dimensions of social values (value provider, value scale, intention, and elicitation process) cases will be developed in order to illustrate the various

potential forms of value expression in an economic environmental valuation and to identify which of these cases are associated with social values.

Table 4-4 illustrates the multiple forms of value expression based on the various combinations of the four characteristics. The resulting value indicator is either disaggregated (individual scale) or aggregated in terms of a larger entity such as groups, communities or societies. Social value expressions are marked bold and are underlined. As indicated by the table, value expressions may be “social” with respect to all four characteristics or just a single one. The various forms of value expression will be discussed in more detail.

The first four cases focus on the individual value provider. The first case (I) describes conventionally elicited WTP. An individual is asked for her WTP under consideration of personal costs and benefits. Assuming a rational individual self-interested utility will be expressed in form of exchange value. The latter can be aggregated in order to express value to society (case II). Hence, the second case has a social scale and reflects social value as conceptualised in neoclassical economics. Further, individuals can be asked how much society should be willing to pay for an environmental good (e.g. how to split a public budget) which without time to reflect or deliberation results in a speculative value (case III) (Sagoff, 1998;

Spash, 2007).

Case IV places individuals behind a veil of ignorance in order to elicit public preferences of impartial citizens. Behind the veil of ignorance, the valuing agents do not know their position in society (Harsanyi, 1953, 1955; Rawls, 1971/2009). Hence, it is assumed that the value indicator reflects a just price based on the principle of justice concerned with distribution. Yet, the principles differ between Harsanyi and Rawls. According to Harsanyi (1955) citizens would maximise utility under uncertainty suggesting that equal probability is given to all outcomes.

Hence, the average utility principle serves as principle of justice. Put another way, an individual is expected to maximise the expected utility of the average citizen. Hence, on a social scale, value is expressed as weighted sum of expected utilities.

Table 4-4 Identifying multiple forms of (social) value expression

Case Provider Scale Intention Elicitation process Value indicator disaggregated aggregated II Social Self-regarding Aggregation procedure

Aggregated individual

WTP III Social Self-regarding Individually stated

preferences VI Individual Self-regarding Individually stated

preferences Source: Own illustration based on Spash (2007, p. 696)

Attributes associated with social values are in bold and underlined.

In contrast, Rawls (1971/2009) argued that in the original position rational behaviour is characterised by application of the maximin rule. According to the latter all alternatives should be ordered based on their worst possible outcomes. The “best” alternative to be chosen is the one which performs best among the worst outcomes. Thereby, an adequate minimum for everyone is secured which according to Rawls is rational when no one knows her position in society nor any probabilities. However, he argues that this outcome is non-utilitarian (Rawls, 1971/2009). No matter which principle of justice is applied, in case IV the outcome is associated with social values due to impartial and impersonal preferences which involve fairness and the social good.

Placing the individual in a group context (cases V to IV), e.g. focus groups, results in different value expressions. In line with preference economisation approaches, after deliberation the participants are expected to be better informed consumers stating well-informed WTP amounts (case V). As discussed in context of complex human behaviour, also altruism can determine behaviour.

Cases VI and VII refer to Sen’s concept of sympathy and commitment (Sen, 1977b). As discussed above, in case that altruism is also linked to self-interest, e.g. due to interdependent preferences, the underlying motivation is sympathy. Thus, the underlying intention may be completely self-interested or somewhere in-between self-interest and other-regarding preferences. Yet, commitment is other-regarding as the WTP is independent of personal utility.

Further, commitment may involve fairness considerations. The point of reference may either be the individual, e.g. choosing a fair amount in terms of ability to contribute to a common good, or societal costs, in which case it can be reasoned what a fair amount for everyone would be.

Although, this type of behaviour would be termed strategic behaviour or rather free-riding in neoclassical economics, it can also be understood in terms of community members’ WTP representing a fair share for a public good (see Sagoff, 1998, p. 226). While preference endogeneity suggests that in the best case a social learning process takes place, this cannot be guaranteed.

In the worst case (case VIII) the individual is negatively influenced by other group members, e.g. through social desirability bias, manipulation or power relations (see e.g. Norwood & Lusk, 2011). Hence, the WTP amounts would be distorted. Also, strategic behaviour can occur in case that individuals observe the behaviour of other group members and choose their WTP in response, e.g. decide to free-ride (see e.g. Bateman et al., 1995; Samuelson, 1954). Lastly,

individuals in group contexts can be asked how much society should be willing to pay for an environmental good (case IX). In contrast to case III, this value is not considered to be speculative anymore as deliberation and time to reflect are part of the elicitation process. Hence, in case IX the resulting aggregated value indicator is expressed social WTP stated by citizens instead of consumers.

Furthermore, groups can express monetary value indicators in form of collective value judgements (cases X to XIII). In case X, the group decides on a fair price an individual should pay, whereas in case XI the value indicator is on a social scale reflecting arbitrated social WTP.

These two cases are associated with impartiality and impersonality. Also in DMV studies these intentions are often linked to Rawls’ theory of justice and liberal democracy (Lo & Spash, 2013). For example, Brown et al. (1995) refer to Harsanyi’s and Rawls’ original position in their values jury approach. Selected members form a values jury in order to obtain public value judgments. To do so, jury members are asked to act as representatives of current and future generations. Again, the underlying assumption is that a social learning process takes place which enables mutual understanding to reach consensus. However, in practical terms consensus is often unrealistic and in line with other supermajority rules, it tends to favour the status quo (Gaus, 2008). Also, majority voting is a two-sided medal. One the one hand, the majority can limit individually rational unsustainable use of environmental resources, on the other hand, it can also result in collective harm in form of support for overexploitation of environmental resources. Further, while majority voting emphasises participation and democracy it may also lead to a trade-off between two democratic values: equality (e.g. preservation of resources for future generations) and freedom (due to restriction of individual consumption) (Edney, 1980).

Still, majority voting is commonly accepted as alternative procedure when consensus fails (see e.g. Howarth & Wilson, 2006; Ward, 1999; Wilson & Howarth, 2002).

Further, as indicated by cases XII and XIII, deliberation may fail the original objective, resulting in diverging value indicators. Before turning to the non-social value judgements in a group setting, first it shall shortly be discussed why deliberation could fail. According to Habermas (1984, p. 285), types of action can be distinguished along two dimensions: action orientation and action situation (see Table 4-5). Action orientation can be oriented towards success or towards reaching understanding. The action situation can either be non-social or social. According to Habermas, instrumental action is always characterised as non-social and directed towards own benefits. In contrast, not every action in context of a social action situation is oriented towards reaching understanding (which represents Habermas’ ideal of

communicative action). Hence, also an action in a social setting can be success oriented which represents strategic action (Habermas, 1984, p. 285). Therefore, also in a group context not every preference must be social.

Table 4-5 Habermas' types of action

Action Orientation

Action Situation Oriented to Success Oriented to Reaching Understanding

Non-social Instrumental action -

Social Strategic Action Communicative Action

Source: Own illustration adapted from Habermas (1984, p. 285)

Turning back to the value expressions identified in Table 4-4, case XII and XIII describe situations in which case a “false” consensus or rather preference convergence is caused by social pressure either due to group dynamics or powerful group members. Hence, the value indicator does not reflect social value but idiosyncratic characteristics of a smaller entity (e.g.

subpart of society, a particular group, or powerful group member(s)) (Dietz et al., 2009; Janis, 1982; Maier, 1967; van Mill, 2007, p. 66). “False” consensus may also occur due to groupthink in which case in-group members aim to reach (unanimity) consensus which is considered a self-serving end and perceived more important than reaching mutual understanding and consensus through the best argument (see Janis, 1972). In line, group polarization may occur in deliberative settings. In case of group polarization deliberation causes groups to take more extreme positions towards opinions they were prone to in the beginning of the process (Brown, 1986, p. 206ff.). Thereby, on an individual scale fair prices may turn into “dictated” prices and on a social scale the group WTP lacks arbitration (here referred to as idiosyncratic group WTP).

As illustrated above, only in one case actual exchange value is measured suggesting that in most of the cases WTP is not an individual welfare measure in a strict neoclassical economic sense.

Quite severely, Brown (1984, p. 244) notes that ‘there is no such thing as the value of an object (...) because assigned value reflects the context in which the valuation took place and the perception and held values of those assigning value’. The same applies to the concept of social value as it was shown that social value expressions or judgements are feasible in completely different valuation contexts.

This contextuality of value (expressions) highlights the need for qualitative analysis of value indicators. Economic methods cannot adequately account for the multidimensionality of

motivations, value pluralism and the associated manifold value indicators. Incorporating psychological factors helps to understand the motives behind stated preferences (Ajzen, 1991;

Spash, 2002).

Yet, even if qualitatively interpretable, quantitative measures obtained in valuation studies may not be comparable between studies and potentially also within studies if respondents adopt varying rationalities. So, even if the value expressions can be distinguished, the question remains how to compare apples (individual preferences) with oranges (social preferences, e.g.

including morals) (see Nyborg, 2000)? The problem is that in case of WTP associated with social values, Hicksian welfare changes cannot be defined because social preferences are not strictly utilitarian nor consequentialist. One possible solution is to adopt a pragmatic, instrumental perspective – to nevertheless compare conceptually different preferences if the value indicator is identical (e.g. monetary). However, while it is possible to do so, it diminishes

including morals) (see Nyborg, 2000)? The problem is that in case of WTP associated with social values, Hicksian welfare changes cannot be defined because social preferences are not strictly utilitarian nor consequentialist. One possible solution is to adopt a pragmatic, instrumental perspective – to nevertheless compare conceptually different preferences if the value indicator is identical (e.g. monetary). However, while it is possible to do so, it diminishes